1. Field of the Invention
The present invention relates to a method of and a crucible for preparing a compound semiconductor crystal, and more particularly, it relates to a method of and an apparatus for preparing a compound semiconductor crystal such as a group III-V compound semiconductor crystal of GaAs, GaP, GaSb, InP, InAs or InSb, or a group II-VI compound semiconductor crystal of CdTe, CdMnTe, CdZnTe, HgCdTe, ZnSe or ZnSSe.
2. Description of the Background Art
Typical conventional methods of preparing compound semiconductor bulk crystals are a horizontal Bridgman method, a gradient freezing method, a vertical Bridgman method, and a vertical gradient freezing method. In order to grow a single crystal by any of these methods, a seed crystal is first arranged on an end of a boat or a crucible and brought into contact with a raw material melt, and the temperature is gradually reduced from the seed crystal side to grow a single crystal. In such a method, however, the crystal is grown while being in contact with the container wall, and hence crystal defects are disadvantageously caused due to wetting of the melt and the container wall.
It has generally been known in the art that crystal defects are caused in wet portions of the melt and the container wall, to result in polycrystallization. In order to solve this problem, the following countermeasure has been considered in general:
A boat or a crucible is generally made of quartz or PBN (pyrolytic boron nitride). In relation to a boat or a crucible of quartz, it is known possible to reduce its wettability by roughening its surface by sandblasting. In relation to a crucible of PBN, on the other hand, it is known possible to prevent wetting by covering its inner surface with boron oxide (B.sub.2 O.sub.3) which is introduced therein with the raw material, for example.
FIG. 5 is a sectional view showing an exemplary apparatus for preparing a GaAs single crystal by the vertical Bridgman method.
Referring to FIG. 5, this apparatus comprises a stainless chamber 1, a heat insulator 2 which is arranged along its wall surface, heaters 3 to 12 which are arranged in its interior, and a quartz ampoule 14, supported by a lower shaft 13, which is mounted on its center. An arsenic pressure control part 15 is provided on a lower end of the ampoule 14 for controlling the temperature of this portion, thereby controlling the arsenic partial pressure in the ampoule 14. The arsenic partial pressure is controlled with solid arsenic 16.
In the apparatus having the aforementioned structure, a GaAs crystal is prepared in the following manner:
First, a crucible 17 is arranged in an upper portion of the ampoule 14. The lower portion of the crucible 17 is in the form of an inverted cone, so that a seed crystal 18 is mounted on its lower end. Then, GaAs raw material 19 and B.sub.2 O.sub.3 are charged on the seed crystal 18 which is arranged in the crucible 17. Thereafter the ampoule 14 is sealed, and heated by the heaters 3 to 12. In this heating, the B.sub.2 O.sub.3 is first softened to cover the inner surface of the crucible 17. Thereafter the temperature is further increased to melt the GaAs raw material 19, and temperature distribution is optimized, so that the lower shaft 13 is thereafter downwardly moved to move the ampoule 14 toward a lower temperature side, and the raw material 19 is solidified from the seed crystal 18 side, thereby growing a single crystal.
The vertical gradient freezing method is carried out substantially similarly to the vertical Bridgman method shown in FIG. 5, except that the vertical position of the ampoule 14 is left intact and outputs of the heaters 3 to 12 are varied to gradually upwardly move the lower temperature portion so that the raw material 19 which is stored in the ampoule 14 is solidified from the seed crystal 18 side for growing a single crystal.
When the vertical Bridgman method or the vertical gradient freezing method is employed, it is possible to alternatively grow a crystal in the atmosphere, without employing the stainless chamber 1. In this case, the heat insulator 2, the heaters 3 to 12 and the like must be formed to be employable in the atmosphere.
In the aforementioned conventional method, however, it is not necessarily possible to completely cover the inner surface of the crucible with B.sub.2 O.sub.3, although the raw material and B.sub.2 O.sub.3 are introduced into the crucible together and the ampoule is sealed so that B.sub.2 O.sub.3 is softened to cover the inner surface of the crucible. To this end, Japanese Patent Laying-Open No. 62-176998 (1987) proposes a method of previously spreading boron oxide powder or boric acid powder on an inner surface of a PBN boat. Further, each of Japanese Patent Laying-Open No. 62-176998 and U.S. Pat. No. 4923561 proposes method of previously heating a PBN boat or a BN crucible in an oxygen atmosphere and oxidizing the same thereby previously forming a B.sub.2 O.sub.3 coat on the inner surface of the boat or the crucible, and thereafter introducing raw material into the same for growing a crystal. According to this method, the inner surface of the boat or the crucible is previously covered with B.sub.2 O.sub.3, and hence it is conceivably possible to attain a higher effect of preventing polycrystallization caused by wetting, as compared with the conventional method of charging B.sub.2 O.sub.3 with the raw material.
However, the aforementioned methods have the following problems respectively:
First, the method of previously spreading boron oxide powder or boric acid powder on the inner surface of a PBN boat or crucible has the following problems:
First, it is scarcely possible to attain adhesion of the boron oxide powder or the boric acid powder to the inner surface of the crucible by simply spreading the same. Therefore, the powder is disadvantageously easily separated from the crucible surface through contact with the raw material as introduced. Consequently, it is extremely difficult to homogeneously form a B.sub.2 O.sub.3 film.
Second, the boron oxide powder or the boric acid powder has a high water content, and water-containing B.sub.2 O.sub.3 is extremely easy to scatter. Therefore, this B.sub.2 O.sub.3 is scattered before melting of the raw material, to break the film and expose the inner surface of the crucible. Consequently, it is extremely difficult to obtain a homogeneous film since B.sub.2 O.sub.3 is remarkably scattered particularly under a vacuum.
On the other hand, the method of previously heating a PBN boat or a BN crucible in an oxygen atmosphere and oxidizing the same for previously forming a B.sub.2 O.sub.3 coat on the inner surface of the boat or the crucible has the following problems:
First, it is generally necessary to form a thick B.sub.2 O.sub.3 film of about 50 .mu.m, in order to effectively prevent polycrystallization caused by wetting. Due to the formation of such a thick B.sub.2 O.sub.3 film, therefore, the thickness of the crucible is remarkably reduced and hence its life is extremely reduced.
Second, it is necessary to heat the inner surface of a crucible which is made of PBN (pyrolytic boron nitride) having a dense structure for a long time in order to oxidize the same, and hence the preparation cost is inevitably increased.
Third, it is difficult to maintain a homogeneous flow of oxygen gas although the thickness of the B.sub.2 O.sub.3 depends on this flow, and B.sub.2 O.sub.3 is reduced in viscosity due to the high temperature heating and collected in the lower portion of the crucible. Thus, it is extremely difficult to form a homogeneous B.sub.2 O.sub.3 film.